Scribd
Upload
37 views1 page

Two Samples T Test Same N

The document explains how to perform a one-tailed t-test for paired samples using Excel, comparing two data sets A and B to determine if there is a statistically significant improvement. It provides calculations for sample means, standard deviations, t-statistic, and p-value, concluding that there is insufficient evidence to claim a significant increase from Sample A to Sample B since the confidence level is below 95%. The document also includes a comparison with Excel's built-in t-test results.

Uploaded by

Jaydee
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLS, PDF, TXT or read online on Scribd
37 views1 page

Two Samples T Test Same N

The document explains how to perform a one-tailed t-test for paired samples using Excel, comparing two data sets A and B to determine if there is a statistically significant improvement. It provides calculations for sample means, standard deviations, t-statistic, and p-value, concluding that there is insufficient evidence to claim a significant increase from Sample A to Sample B since the confidence level is below 95%. The document also includes a comparison with Excel's built-in t-test results.

Uploaded by

Jaydee
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLS, PDF, TXT or read online on Scribd
You are on page 1/ 1

Paired Samples - Using Excel's t-test for Paired Samples

[This example is for a one-tail hypothesis test]


J. M. Cimbala
Given: Two data sets, A and B of the same variable.
Goal: Determine whether there is a statistically significant improvement in the variable.

data # xA xB d = xB - x A
1 25.6 26.2 0.6 Calculate the sample mean and
2 27.3 27.1 -0.2 sample standard deviation of d:
3 24.2 24.1 -0.1
4 28.7 29.2 0.5 dmean = 0.25
5 23.6 24.5 0.9 Sd = 0.476445
6 25.1 24.9 -0.2 n= 6
Sample means: 25.75 26 0.25 df = 5
Sample standard deviations: 1.927433527 1.92665513 0.47644516998

Null hypothesis: Sample B has a smaller mean than Sample A


OR, mathematically, md < 0 (population mean of variable d)
m0 < 0

Calculate the t-statistic and the p-value:


(Note: Always use absolute value here for the t-statistic)
t-statistic: t = 1.28529467 [ t = ABS((dmean - m0) * SQRT(n)/Sd ) ]
number of tails = 1 [This is a one-sided hypothesis test.]
p-value: p = 0.12750083 [ p = TDIST(t,df,tails) ]
Conclusions: There is a 12.7500825 % probability that the null hypothesis is correct.
There is a 87.2499175 % probability that the null hypothesis is not correct.
We are confident to 87.25 % that there is an increase in the variable.

Now let's repeat the analysis using Excel's built-in t-test: Conclusion:
(Data Analysis-t-Test: Paired Two Sample for Means) Since the confidence level is < 95%,
we cannot claim a statistically significant
t-Test: Paired Two Sample for Means improvement (increase in the mean)
from Sample A to Sample B.
Variable 1 Variable 2
Mean 25.75 26
Variance 3.715 3.712
Observations 6 6
Pearson Correlation 0.969435921
Hypothesized Mean Difference 0
df 5
t Stat -1.28529467 This is our t-statistic
P(T<=t) one-tail 0.127500825 This is our p-value
t Critical one-tail 2.015048372 This is the critical value of t (ta/2) for 95% confidence
P(T<=t) two-tail 0.25500165 Ignore these values for our one-tail test (these values are for a two-tail test)
t Critical two-tail 2.570581835 Ignore these values for our one-tail test (these values are for a two-tail test)
Verify the critical t value for 95% confidence: ta/2 = 2.015048 [ ta/2 = TINV(0.10,df) ]
Note: For 95% confidence with one tail, a/2 = 0.05, so a = 0.10.

You might also like